Differential Gene Expression Patterns and Interaction Networks in BCR/ABL Positive and Negative Adult Acute Lymphoblastic Leukemias.

BCR/ABL is associated with an unfavorable prognosis in adults with acute lymphoblastic leukemia (ALL). We used DNA microarrays to identify gene expression profiles and molecular interaction networks related to BCR/ABL status and clinical outcome in a set of 54 adult ALL specimens from the MRC UKALL XII/ECOG E2993 intergroup study (21 p185^BCR/ABL- and 16 p210^BCR/ABL-positive and 17 BCR/ABL negative). In order to avoid biases associated with commonly used sample amplification procedures, we have implemented an indirect two-step labeling protocol based on signal amplification by use of dendrimer technology. Using a two-class, non-parametric t-test and a false discovery rate cutoff of 5%, we identified 271 genes (including GAB1, CIITA, XBP1, CD83, SERPINB9, PTP4A3, NOV, LOX, CTNND1, BAALC and RAB21) as differentially expressed in BCR/ABL positive ALL compared with BCR/ABL negative ALL. They separate these two classes of adult ALL with an overall accuracy of 93% and are enriched for three highly relevant biological functions: Cellular Growth and Proliferation (57 genes, p = 0.004–0.044), Cell Death (49 genes, p = 0.0007–0.049), and Hematological System Development and Function (40 genes, p = 0.00004–0.049). Network analysis demonstrated complex interaction patterns of these genes, and identified FYN and IL15 as the hubs of the top-scoring network. We confirmed these findings by both qRT-PCR analysis of the initial set of samples and by cross-platform validation in an independent cohort of 128 adult ALL specimens. In addition, within the BCR/ABL positive subgroup, we identified 14 clones found to be over-expressed (TSPAN16, ADAMTSL4) or under-expressed (PILRB, STS-1, SPRY1) in p185^BCR-ABL- relative to p210^BCR-ABL-ALL. In a nearest-centroid classification, these clones correctly predict the BCR/ABL subtype with a cross-validated prediction accuracy of 95%. No differential gene expression was detected among Rho family GTPases and their known interaction partners. Finally, we constructed a gene expression- and interaction-based outcome predictor consisting of 27 genes (including GRB2, GAB1, GLI1, IRS1, RUNX2 and SPP1), which strongly correlated with overall survival in BCR/ABL positive adult ALL (p=0.0001), independently of other clinical parameters such as age (p=0.25) and white blood cell count at presentation (p = 0.003). These findings may be useful for developing novel therapeutic targets and prognostic markers in adult ALL.